Experimental study of steel plate shear walls with infill plates strengthened by GFRP laminates

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Experimental study of steel plate shear walls with infill plates strengthened byGFRP laminates

F. Nateghi-Alahi, M. Khazaei-Poul ⁎Structural Engineering Research Center, International Institute of Earthquake Engineering and Seismology (IIEES), 19395 Tehran, Iran

a b s t r a c ta r t i c l e i n f o

Article history:Received 24 November 2011Accepted 2 July 2012Available online xxxx

Keywords:CompositeSteel plate shear wallGFRP laminateFiber orientationSeismic behaviorStiffnessFailure modes

In this paper, the nonlinear behavior of the composite steel plate shear walls, in which steel infill plates wasstrengthened by the fiber reinforced polymers (FRP), is experimentally investigated. Tests are designed to eval-uate the effect of Glass-FRP layers, the number of GFRP layers and the orientation of GFRP layers on the stiffness,shear strength, cumulative dissipated energy and other major seismic parameters in the composite steel plateshear walls. Experimental models are scaled as one-story steel shear panel models with hinge type connectionsas boundaries at four corners. In thefirst test, unstiffened steel infill plate is used for the testing as a reference test.However, in the next four tests, strengthened steel infill plateswere usedwith different numbers and orientationof GFRP layers. Each testwas performedunder fully reversed cyclic quasi-static loading in the elastic and inelasticresponse zones of the specimens, in compliance with the ATC-24 (1992) test protocol. The experimental resultsindicated that by strengthening the infill steel plates yield by laminates strength, ultimate shear strength and cu-mulative dissipated energy can be significantly increased. The paper therefore presents the background, proce-dure and set-up used as well as the test results.

© 2012 Elsevier Ltd. All rights reserved.

1. Introduction and background

Steel plate shear walls (SPSWs) possess significant advantages overmany other systems in terms of cost, substantial ductility, high initialstiffness, fast pace of construction, and the reduction in seismic mass[1]. The SPSW system can be used in different configurations, such as:

a. Unstiffened steel plate shear wall;b. Stiffened steel plate shear wall;c. Composite steel plate shear wall.

Unstiffened-SPSWs are the basis for the SPSW systems. Unstiffenedweb plate has negligible compression strength; therefore shear buck-ling occurs at the low levels of loading. Lateral loads are then resistedthrough diagonal tension in the web plate. In addition, stiffened webmay also be used to increase the shear buckling strength. In this typeof SPSWs, the shear strength is the result of a combination of shearbuckling strength and the additional strength from the diagonal tensionfield action [2]. In Composite Steel Plate ShearWalls (C-SPSW), the steelweb plate can be stiffened by adding concrete on one or both sides ofthe web plate. Concrete layers can improve the load carrying capacityof the SPSWs by permitting the utilization of the full yield strength ofthe infill plates. In addition, the shear strength of the concrete is effec-tive in increasing the capacity of the system all together [1].

FRP laminates are high strength, high stiffness, light weight, flexibil-ity to form in any shapes, easy to handle during construction, and excel-lent resistance to corrosion and environmental degradation. Thesesuperior mechanical properties of the FRP laminates have made theman effective alternative for steel plates used for strengthening andupgrading steel structures [3,4]. Steel infill plates can be strengthenedby adding a number of layer of FRP laminates in both sides. In thistype of C-SPSW, like unstiffened SPSW systems, the strengthenedsteel plate has negligible compression strength and shear bucklingoccurs at low levels of loading cycle. Lateral load is resisted throughdiagonal tension in the web plate. In C-SPSW, FRP laminate is effectiveto increase post‐buckling strength, initial and secant stiffness of thesystem as well [5].

During the last four decades many experimental and numericalstudies on seismic performance of SPSWs have been carried out. Allthese works have led to the better understanding of this lateral load re-sistant system. Wagner [6] was the first researcher who used completeand uniform tension fields to determine the shear strength of a panelwith rigid flanges and very thin webs, and inferred that the shear buck-ling of a thin aluminum plate supported adequately on its edges doesnot constitute failure. Other researchers conducted works based onthis idea in order to develop analytical methods for modeling andanalysis of thin SPSWs. Thorburn et al. [7] developed a simple analyticalmethod to evaluate the shear strength of unstiffened SPSWs with thinsteel plates and introduced the strip model to represent the tensionfield action of a thin steel wall subjected to shear forces. Timler andKulak [8]modified the formula for the angle of inclination of the tensionstrips by the tests. Elgaaly [9] experimentally investigated the behavior

Journal of Constructional Steel Research 78 (2012) 159–172

⁎ Corresponding author. Tel.: +98 9386012122.E-mail addresses: [email protected] (F. Nateghi-Alahi), [email protected],

[email protected] (M. Khazaei-Poul).

0143-974X/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.doi:10.1016/j.jcsr.2012.07.002

Contents lists available at SciVerse ScienceDirect

Journal of Constructional Steel Research


of SPSWs and proposed an analytical model to determine the behaviorof thin steel plate shear walls. Berman and Bruneau [10] presentedplastic analysis method based on the strip models as an alternative forthe design of SPSWs. Strip model has been implemented into theCanadian design codes (CAN/CSA 2001) [11] and the AISC seismic de-sign specifications [12]. Sabouri-Gohomi et al. [13] proposed Plate–Frame-Interaction (PFI) method to predict the shear behavior of theSPSWs. Kharrazi and et al. [14] presented the Modified Plate Frame In-teraction (M-PFI) method for the use in the design of steel plate wallsystems. Khazaei-Poul and Nategh-Alahi [15] proposed an analyticalmodel, the Composite–Plate Frame Interaction (C-PFI) method, to pre-dict the shear behavior of the SPSWs with infill plates strengthenedby the GFRP laminates. They also showed that the C-PFI method isable to properly predict the shear behavior of the C-SPSW systems.

Astaneh-Asl [16,17] performed experimental tests on two speci-mens of a three-story C-SPSW under cyclic loads. He showed that theconcrete layer produces better distribution of stresses in the steelplate and develops tension field lines in a wider region. Rahai andHatami [18] experimentally and numerically investigated the effectsof the shear stud spacing variation, middle beam rigidity and the meth-od of beam to column connection on the C-SPSWs behavior.

Lubell et al. [19] tested two single and one 4-story thin SPSWS undercyclic loading and compared the experimental results with the simpli-fied tension field analytical models and found that the models can pre-dict post-yield strength of the specimens well, with less satisfactory inthe elastic stiffness results. Caccese et al. [20] tested five one-fourthscale models of a three-story into the effects of the panel slender ratioand type of the beam-to-column connections. They reported that asthe plate thickness increased, the failuremodewas governed by columninstability and the difference between simple and moment-resistingbeam-to- column connection was small. Driver et al. [21] tested a4-story large-scale SPSWspecimenwith unstiffened panels under cyclicloading to determine its behavior under an idealized severe earthquakeevent. Robert and Sabouri-Gohomi [22] conducted a series of 16quasi-static loading tests on unstiffened steel plate shear panels withcentral opening. Vian et al. [23] performed test on the special perforatedSPSWs with reduced beam section under cyclic loading and reportedthe perforated panel reduced the elastic stiffness and overall strengthof the specimen by 15% as compared with the solid panel specimen.

Alinia [24] and Alinia and Dastfan [25] studied the effect of thesurrounding members on the overall behavior of the thin steel plateshear walls. His results show that the flexural stiffness of the sur-rounding members has no significant effect on the elastic shear buck-ling or the post-buckling behavior of the shear walls.

Also, several studies on the FRP strengthened of steel structures havebeen carried out and reported in literature. The “CIRIA Design Guide”[26] provides detailed design guidance for strengthening metallic struc-tures using externally bonded FRP. Wright et al. [27] experimentally in-vestigated a joint between steel plates and FRP laminate and found that,by increasing the thickness of the adhesive layer, the relative stress con-centration level was reduced by 21%. Ekiz and El-Tawil [28] carried outan analytical and experimental research program in order to investigatethe buckling behavior of the compressive steel braces strengthenedwithCFRP laminates. Sen et al. [29] studied the strengthening steel bridgesections using CFRP laminates. Moy et al. [30] reported the upgrade ofthe cast iron beams on the London underground system by bondingCFRP plates to the tension flanges. Deng and et al. [31] investigated thebehavior under static loading of the metallic beams reinforced withbonded CFRP plates. Benachour and et al. [32] developed a new solutionto predict both shear and normal interfacial stresses in the simplysupported beams strengthened by bonded pre-stressed composite lam-inates. Studies on the fatigue crack propagation in steel membersstrengthened by the CFRP were also carried out by [33–35]. Miller etal. tested two full-scale steel-CFRP composite girders under fatigueloads [36]. Hollaway and Cadei reported an excellent state-of-the art re-view on the problems encountered in plate bonding on to the metallic

structures and discuss how these problems might be overcome [37].Zhao and Zhang wrote a comprehensive review article on the bond be-tween steel and FRP, the strengthening of steel hollow sectionmembers,and fatigue crack propagation in the FRP–Steel systems [4].

As seen in by the review of the literature, it seems that these areneeded to investigate the composite behavior of steel plate and FRPlaminates. Therefore the main purpose of this study was to investigatethe combined behavior of both the steel plates and FRP laminates to-gether and determine the enhanced behavior if any. This method ifproved to be positive, could therefore be used in existing systems forenhancing the behavior, such as existing steel shear walls with aknown thickness, ship manufactures, aerospace people for repair andmanufacture purposes andmany other potential applications. As statedthemain purposewas to investigate the composite behavior of the steeland the FRP laminate under shear loading.

In this study, the nonlinear behavior of strengthened steel plateshear walls by GFRP laminate has been experimentally investigated. Inthis regards, five tests were conducted. Tests were designed to evaluatethe effect of GFRP layers, number of GFRP layers and the orientation ofGFRP layers on the stiffness, shear strength, ductility, cumulative dissi-pated energy and other important seismic parameters. Experimentalmodels are scaled as a one-story steel shear panels, with hinge typeconnections of the boundaries at four corners. In the four tests, steelinfill plates were strengthened by GFRP laminates and all specimenswere subjected to the quasi-static loadings. The shear capacities andthe hysteresis curves of the experimental unstiffened and strengthenedsteel plate shear panels were then compared. The experimental resultsindicated that with strengthening steel infill plate, yield strength, theultimate shear strength and the cumulative dissipated energy can besignificantly increased.

2. Shear load–displacement diagram of composite steel plateshear wall

In SPSWs, steel infill plates can be strengthened by a number of FRPlaminated layers under different orientations, (see Fig. 1). In Fig. 1, the(X–Y) coordinate system is the general coordinate system and (1, 2) isthe principal material coordinates of a typical FRP laminated layer.The (X–Y) coordinate system in Fig. 1 does not coincidewith the princi-pal material coordinate system. The 1-axis is oriented at an angle of +γcounterclockwise from the X-axis. In (X′–Y′) coordinate system, X′-axisis oriented along the direction of the tension field and at an angle of +θcounterclockwise from the X-axis. Khazaei-Poul and Nateghi [15]numerically investigated the behavior of steel plate shear walls having

Fig. 1. General configuration of the composite steel plate shear wall.

160 F. Nateghi-Alahi, M. Khazaei-Poul / Journal of Constructional Steel Research 78 (2012) 159–172


infill plates strengthened by GFRP laminates and they showed that fiberorientation is an important variable on the seismic behavior of theC-SPSWs. They showed that if principal orientation of GFRP laminatesis oriented in the direction of tension fields, the shear strength and stiff-ness of C-SPSW can increase.

FRP layers can be attached to steel infill plates by the use of adhe-sive. If a perfect bond is achieved at the bond between adhesive andsteel infill plate interface and FRP layers and considering that no de-lamination occurs in the FRP layers, shear load–displacement diagramof the composite steel plate shear wall can be approximately obtainedby superimposing the shear load–displacement diagrams of the frameand the composite steel plate as shown in Fig. 2 [15]. In C-SPSWs afteryielding of the steel infill plate, the system will have a secant stiffnessthat it is provided by the FRP laminate layers.

The shear load–displacement diagram of the FRP layers only, steelplate only, and composite steel plate with height d, width b, and thick-ness t (see Fig. 3) is shown in Fig. 4. The behavior of the FRP layers is as-sumed approximately linear until the failure and steel plate haselastic-perfect-plastic behavior. In this figure, point A corresponds tothe buckling limit of the composite steel plate, and point B correspondsto the yielding point of the steel plate in composite steel plate. Point Ccorresponds to the displacements larger than the displacement atpoint B, where steel plate is yielded.

Shear strength of the composite steel plate at the yield of the steelplate is the combination of the shear buckling strength and the addi-tional strength from the diagonal tension action. However, the bucklingstrength of the composite steel plate in comparison to the post-bucklingstrength of it is insignificant and low. Limiting the elastic shear strengthof the composite steel plate, FCSPy, by ignoring the buckling componentdue to the formation of tension field lines is a combination of the

post-buckling strength of the steel plate layer and the FRP laminatedlayers, and it can be calculated as [15]:

FCSP:y ¼ Fsteel plate þ FFRP layers ¼12:Fy:b:tst: sin2θ

þXNi¼1

12: σ lami

¼ Ex′ iαi:Est

Fy

� �:b:tlami

: sin2θ

ð1Þ

After yielding of the steel plate layer, the secant stiffness of thecomposite steel plate is provided by the FRP laminated layers.

The equivalent stiffness of the composite steel plate, Kcomp.plate.sec,is then given by [15]:

Kcomp:plate: sec≅XNi¼1

Ex′ i :b:tlami: sin2θð Þ2

4:dð2Þ

The total shear strength of the composite steel plate before failureof the FRP layers, FCSPΔ, is given by:

FCSP:Δ ¼ FCSP:y þ Kcomp:plate: sec � Δ−UCSP:y

� �¼

¼ 12:Fy:b:tst : sin2θþ

XNi¼1

12: σ lami

¼ Ex′ iαi:Est

Fy

� �:b:tlami

: sin2θ

þXNi¼1

Ex′ i :b:tlami: sin2θð Þ2

4:d� Δ−UCSP:y

� �

Notationsb Width of paneld Height of paneltst Thickness of steel infill platetlami Thickness of i-th FRP laminate layerθ Inclination angle of tension fieldN Number of FRP laminate layersαi Factor between the strain in i-th layer of FRP laminate and

steel infill plateEst Elasticity modulus of steel infill plateEXi Elasticity modules of i-th layer of the FRP laminate layers in

the direction of tension field linesUCSP.cr Critical shear displacement of composite steel plateUCSP.y The limiting elastic shear displacement of composite steel

plateUFry The limiting elastic shear displacement of the frameD Displacements larger than yielding point of composite steel

plate

Fig. 2. Shear load–displacement diagram of C-SPSW [15].

Fig. 3. Composite plate–frame interaction (C-PFI) model plate idealization [15].

Fig. 4. Shear load–displacement diagram of composite steel plate only, steel plate onlyand FRP laminate layers only [15].

161F. Nateghi-Alahi, M. Khazaei-Poul / Journal of Constructional Steel Research 78 (2012) 159–172


FCSPSW.UCSP.cr Total shear strength of the CSPSWs at UCSP.cr

FCSPSW.UCSP.y Total shear strength of the CSPSWs at UCSP.y

FCSPSW.UFr.y Total shear strength of the CSPSWs at Ufry

FCSPSW.Δ Total shear strength of the CSPSWs at DFCSPcr The critical shear strength of composite steel plate (correspond

to the buckling limit)FCSPy Limiting elastic shear strength of composite steel plateFCSPΔ Total shear force of the composite steel plate at Δ (Δ>UCSP.y)FFr.y The shear strength of the frameFwpb Shear strength of the composite steel plate due to formation

of tension field linesKcomp.plate.sec Equivalent secant stiffness of the composite steel plateKfr Stiffness of frameKcomp.plate Stiffness of the composite steel plateKw Stiffness of steel plateKeq.lam Equivalent stiffness of FRP laminate layers

3. Experimental study

Experimental study was begun by performing several tests. Toinvestigate the effect of GFRP laminated layers on the seismic behaviorof the steel plate shearwalls, several testswere performed on the scaledmodels of the steel plate shear panels. For this purpose, five testswere conducted. Experimental models were scaled as a one-storysteel shear panel models, with hinge type connections in boundariesat four corners. In the first test, the unstiffened steel infill plate wasused for the testing as a reference test. In the next tests, steel infill plateswere strengthened by the number of GFRP laminated layers in different

orientations. In this study, infill plate was strengthened by four patternsin arrangement of the FRP laminates on the infill plates. Each test wasperformed under fully reversed cyclic quasi-static loading in the elasticand inelastic response zones of the specimens in compliance with theATC-24 (1992) test protocol by means of a hydraulic jack of 600 kNcapacity.

3.1. Experimental models and test set-up

For experimental study, the number of cyclic loading tests on asmall-scale, unstiffened steel plate shear panels (SPSP) and the compos-ite steel plate shear panels (CSPSP) were conducted. A scaled one-storysteel shear panel model, with hinge type connections at the boundariesin four corners, was selected. Details of the experimental specimens areshown in Fig. 5.

The edges of the steel and composite plates were clamped betweenpairs of a rigid frame member by the means of two rows of high tensilebolts. Boltswith a diameter of 10 mmwere used for connections of infillplate to the surrounding frame. The boundary elements of the specimenare similar and consisted of the standard profile double sectionUNP100.The boundary elementswere also designed in order tomeet the prelim-inary requirements of the SPSWs andAISC 341‐05provisions. The depthand width of the all specimen were equal to 600 mm while the depthand width of the infill plates were equal to 400 mm. A number of linearvariable displacement transducers (LVDTs) and strain gauges wereplaced on the specimens for measuring the global and local responsesat a specific location of interest strain. A schematic view of the testset-up is shown in Fig. 6. For each test, after preparing, the specimenwas placed in a Roell–Amsler servo-hydraulic testing machine andsubjected to quasi-static cyclic loading along their diagonal axes.

Fig. 5. Details of the experimental specimens (mm).



Properties of the infill plates in the experimental specimens werepresented in Table 1. In the first experimental specimen (SPSP1),unstiffened steel plate with a thickness of 0.9 mm was selected as aninfill plate. In the next models, steel infill plates with the thickness of0.9 mm were strengthened by the numbers of GFRP laminated layersin different orientations that are shown in Fig. 7. For strengtheningthe steel structures, CFRP have some disadvantages in comparison tothe GFRP. CFRP is generally corrosion resistant, but if it is in contactwith the metals, galvanic interaction between the two materials cantake place [38]. The use of GFRP has been proved to prevent this galvan-ic corrosion and to achieve higher bonding strength. In addition, thefailure strain of the GFRP laminates are more than the CFRP laminates.

During testing, the SikaWrap-Hex-430G was applied to strengthensteel plates in all experimental specimens. SikaWrap-Hex-430G is aunidirectional glass fiber fabric. The structural adhesive Sikadur-330was used to bond the composite overlays. The method of applicationby using a grooved roller ensured that the adhesive dispersed uniformlythroughout the fibers. Consequently, a thin layer of epoxy adhesive wasformed at the interface between steel and GFRP layers therefore, work-ing togetherwith the epoxy in-between the fiber laminates. Total thick-ness of eachGFRP laminate layerwith adhesivewas 0.508 mmbased ontechnical data provided by the manufacturer.

Fiber-orientation angle plays an important role in the increase of theplate strength. This is due to FRP having high strength in the fiber direc-tion, and low strength in the direction perpendicular to the fiber. Opti-mum performance of the longitudinal fibers can be obtained if the loadis applied along its direction. The slightest shift in the angle of loadingmay drastically reduce the strength of the composite.

Composite steel infill plate in the CSPSP2 and CSPSP3 specimensare consisted of steel plates with a thickness of 0.9 mm that arestrengthened by one layer of GFRP laminate in each side. In thesespecimens, the total thickness of the composite steel infill plate is ap-proximately equal to 1.916 mm. In the CSPSP2 specimen, principalorientation of GFRP laminates (the direction at which the laminatehas maximum amount of strength and young modules) are orientedat +45° and −45° inclination with angle of tension fields as shownin Fig. 7-a. In the CSPSP3 specimen, principal orientation of GFRPlaminates are oriented in the direction of tension fields as shown inFig. 7-b. Composite steel infill plates in the CSPSP4 and CSPSP5 spec-imens are consisted of steel infill plates with thickness of 0.9 mmwhich are strengthened by two layers of GFRP laminate at eachside. In these specimens, total thicknesses of the composite steelinfill plate approximately are equal to 2.932 mm. In the CSPSP4 spec-imen, like CSPSP2 specimen, principal orientation of GFRP laminatesare oriented at +45° and −45° inclination with angle of tensionfields as shown in Fig. 7-c. In the CSPSP5 specimen, like CSPSP3 spec-imen, principal orientation of GFRP laminates are oriented in direc-tion of tension fields Fig. 7-d. The view of SPSP1 and CSPSP2specimens before test is shown in Fig. 8.

3.2. Material properties

Tests were conducted to determine the stress–strain curves of theinfill steel plates and boundary element materials. All tests wereconducted on specimens dimensioned as proposed in ASTM-A370-05(Standard Test Methods and Definitions for Mechanical Testing of

Fig. 6. Test set-up.

Table 1Properties of infill plate in the experimental specimens.

Experimental models Number of layers incomposite infill plate

Thickness of laminate andsteel infill plate

Total thickness of infill plate Orientation of GFRP layers GFRP type

Steel plate GFRP layer Steel plate GFRP layer

SPSP1 1 0 0.9 mm – 0.9 mm – –

CSPSP2 1 2 0.9 mm 0.508 mm 1.916 mm 0 # 90 SikaWrap® Hex 430GCSPSP3 1 2 0.9 mm 0.508 mm 1.916 mm 45 # −45 SikaWrap® Hex 430GCSPSP4 1 4 0.9 mm 0.508 mm 2.932 mm 0 and 90 # 0 and 90 SikaWrap® Hex 430GCSPSP5 1 4 0.9 mm 0.508 mm 2.932 mm 45 and −45 # 45 and −45 SikaWrap® Hex 430G



Steel Products) [39]. Yield stress and young module of steel infill platebased on the mean of coupon tests are equal to 197 MPa and 204 GPa,respectively and for boundary elements (UNP100) those are equal to310 MPa and 203 GPa, respectively. The failure strains were approxi-mately at 32% for steel infill plate and 19% for UNP100. A summary ofthe test results are presented in Table 2.

Steel infill plateswere strengthened by SikaWrap-Hex-430G. Unlikemost materials, FRP-laminates approximately have almost elastic(linear) behavior untilfinal failure. Based on the technical data providedby the manufacturer, the longitudinal modulus (Ex) and the longitudi-nal tensile strength (Tx) of the cured SikaWrap-Hex-430G Laminatewith Sikadur-330 Epoxy, which is dominated by the properties of thefiber, are equal to 26.49 GPa and 537 MPa, respectively. The transversemodulus (Ey) and the transverse tensile strength (Ty) perpendicularare equal to 7.07 GPa and 23 MPa. The longitudinal and transverse fail-ure strains of the GFRP laminate approximately are equal to 2.2% and0.46%, respectively. Summary of the mechanical properties of theGFRP laminates are presented in Table 3. The Young'smodulus and Ten-sile Strength of Sikadur-330 are 3489 MPa and 60.6 MPa, respectively.

Comparison between strain–stress curves of the steel infill plate andthe GFRP laminated layer in the main direction of the fibers is shownin Fig. 9.

3.3. Composite steel infill plate preparation

To prepare the composite steel infill plates before testing, steel infillplateswere sandblasted in order to remove any rust or grease, oil and toachieve a rough and clean, chemically active surface. In the next step, inorder to remove any dirt, the surface was cleaned using alcohol metha-nol acid. After preparing the steel infill plates, the adhesive was appliedto the steel plate. The mixed Sikadur 330 epoxy resin was applied di-rectly onto the infill steel plate. The SikaWrap Hex 430G laminate wascarefully placed into the applied resin with gloved hands while it wassmoothed out. Also any irregularities or air pockets were worked outwith a plastic laminating roller while letting the resin squeeze out be-tween the rovings of the fabric. A hardening period of two weeksprior to the testing was allowed in order to get the adhesive reach itsmaximum strength.

a) CSPSP2: one layer of GFRP laminate in each b) CSPSP3: one layer of GFRP laminate in each side of steel infill plate; α= +45 & -45

c) CSPSP4: two layers of GFRP laminate in d) CSPSP5: two layers of GFRP laminate in each side of steel infill plate; α= +45 & -45

GFRP Laminate

Steel infill Plate

side of steel infill plate; α= 0 & 90

each side of steel infill plate; α= 0 & 90

Fig. 7. Different patterns used for strengthening of the steel infill plates by GFRP layers.

a) SPSP1 b) CSPSP2

Fig. 8. View of the SPSP1 and CSPSP2 specimens before testing.



3.4. Test loading protocol

The specimens were subjected to a quasi-static cyclic loading alongtheir diagonal axes, which began with very small values of the overalldrift and increased gradually until failure in compliance with ATC-24(1992) test protocol [40]. In Fig. 10-a the load direction is shown onthe specimens. Forceswere applied by using a Schenk 600 kN, servo hy-draulic, dynamic testingmachine. A similar load patternwas used for alltests. In Fig. 10-b, the load protocol that was applied on all the speci-mens is shown.

Before testing, the yield–displacement and shear strength of thespecimens were predicted by the finite element software (ANSYS).The numerical method has been validated using the available experi-mental results in the literature; therefore the SPSW1 specimen ofAlavi-Nateghi's tests [42], Fig. 11, was selected and modeled. TheSPSW1 is ½ scaled one-story specimen with 2 m width and 1.5 mheight of un-stiffened SPSWs. Details of the specimen are presentedin Fig. 11. In the finite element models, SHELL-181 was used formodeling of the steel infill plate and boundary elements. Kinematichardening plasticity model was utilized with multi-linear kinematichardening material model for the mild steel material. The numericalhysteretic and push-over load–displacement curves from the non-linearfinite element analysis are presented and comparedwith the experimen-tal results in Fig. 12. It is obtained that the used numerical method hasbeen successful in estimating the actual shear capacity of the systemand also initial stiffness of the SPSW1, in comparison to the experimentalresults. The difference between obtained shear capacity in the numericaland experimental model is less than 5%.

Numerical push-over curve of the SPSP1 and CSPSP3 specimens arepresented in Fig. 13. Both boundary elements (beam and column) andinfill plate were modeled by using nonlinear shell element (shell-181).SHELL181 is a 4-noded 3-D element with 6 degrees of freedom at eachnode. The element possesses full nonlinear capabilities including largestrain and allows defining 255 layers. The shell section is used formodeling the composite infill plate. Two elements of the MPC184 andCOMBIN7 were used to create hinge connections. The optimum dimen-sion of the elements by trial and error was selected, under the conditionthat the smaller mesh size did not change the results of the analysis. Inthe analysis, the multi-linear kinematic hardening model was assignedto the boundary elements and steel infill plate, while the GFRP layerswere modeled as the orthotropic material. Tsai-Wu Failure Criterion,which allows nine failure stresses and three additional couplingcoefficients, was assigned to the GFRP layers. The principal stresses ofSPSP1 and CSPSP3 are presented in Fig. 14.

3.5. Discussion of the experimental results

3.5.1. SPSP1 specimenThe hysteresis curves for SPSP1 are shown in Fig. 15. In the early

(elastic) loading cycles, the panel behaves in a stiff manner. As the de-formations increase, portions of the steel infill plate yield, resulting ina gradually reducing stiffness. Local buckling after deformations larg-er than 0.42% of the drift could be clearly seen in the steel infill plate.The ultimate strength of the specimen was obtained as 82.58 kN, asshown in Fig. 15. The number of the local tears was initiated by lowcycle fatigue in the non-linear region, resulting from the cyclickinking of the thin steel infill plate at the middle and the corners ofthe panel, Fig. 16. However such tears did not possess significant ef-fect on the shear capacity of the specimen.

Table 2Mechanical properties of the steel materials.

Steel material Elasticmodules

Yieldstress

Yieldstrain

Ultimatestress

Ultimatestrain

Rupturestrain

GPa MPa % MPa % %

Infill plate(thickness=0.9 mm)

204 197 0.097 323.2 25 26

UNP100 203 310 0.23 460.5 18 19

Table 3Mechanical properties of the cured laminate properties of SikaWrap® Hex 430G withSikadur 330 Epoxy.

Tensile modulus Tensile strength

Ex (GPa) Ey (GPa) Tx (MPa) Ty(MPa)

26.49 7.07 537 23

0 1 2 3 40

100

200

300

400

500

Strain (%)

Stre

ss (

MP

a)

Steel Plate t=0.9 mmGFRP layer (main direction)

Fig. 9. Comparison between strain–stress curves of steel infill plate and GFRP laminat-ed layer in the direction of the fibers.

Fig. 10. Load protocol and load direction on the experimental specimen.



3.5.2. CSPSP2 specimenIn the CSPSP2 specimen, the composite steel plate was used as the

infill plate. In this specimen, the principal orientations of the GFRP lam-inates were oriented at +45° and −45° inclination with the angle oftension fields. Hysteresis curves for the CSPSP2 specimen are presentedin Fig. 17. Up to approximately 3 mm of the drift, the response of theCSPSP2 was almost elastic. At deformations larger than 5.25 mm ofthe drift (0.75% drift), with increasing overall deformation, buckling oc-curred in the infill plate. The ultimate shear strength of the CSPSP2 spec-imen at the 7 mm drift level was obtained 12 kN, as shown in Fig. 17.After pushing the drift to the 8 mm level, tears in the GFRP laminatedlayers were gradually expanded and the capacity of the specimen wasreduced. Tears in the direction perpendicular to the main direction ofthe GFRP laminates occurred and as a result, local deboning betweenthe steel infill plate and GFRP laminates took place, as shown inFig. 18. Up to 15 mm drift (2.15% drift), the shear strength of the spec-imen was almost equal to 100 kN and in the larger deformation, itreached up to 80 kN, which is approximately equal to the capacity ofthe SPSP1 specimen.

3.5.3. CSPSP3 specimenThe hysteresis curves for the CSPSP3 specimen are shown Fig. 19. In

the early loading cycles, the panel behaves in a stiff manner. As the de-formations increased, stiffness of the specimen was gradually reduced.Local buckling after deformations larger than 0.72% of the drift observedin the composite infill plate. After yielding of the steel plate (that ap-proximately occurred at 4.4 mm of the drift), due to the GFRP laminateaction, the specimen still behaved in a stiff manner. At the deformationslarger than 6.5 mm, along the direction perpendicular to the main di-rection of the GFRP laminates, tears started showing. The specimen be-haved stiffly up to 13.5 mmof the drift (1.95%drift). At this deformationlevel, the ultimate strength of the SPSP1 specimen was obtained140.8 kN. At the larger deformation levels, because of the growing ofthe delamination, tearing in the GFRP laminated layers and the detach-ment of the bond between steel plate and GFRP laminated layers, grad-ually reduced the shear strength of the specimen. Delamination,detachment and the tearing in the GFRP laminated layer are shown inFig. 20. These results show that the strengthened steel plate has negligi-ble compression strength and shear buckling occurs at the low levels ofthe loading like unstiffened specimen (SPSP1 specimen). In turn, FRPlaminated layers are effective in increasing the post‐buckling shearstrength due to the tension field action and also in providing the addi-tional secant stiffness in the system.

3.5.4. CSPSP4 specimenExperimental results of the CSPSP2 and CSPSP3 specimens, in which

steel infill plates are strengthened by one layer of GFRP at each sideshow that the failure modes of the specimens were consideration ontearing and delamination of the GFRP laminates along the direction per-pendicular to themain direction of the GFRP laminates. Regarding theseproblems and to prevent these failure modes, in the CSPSP4 specimen,steel infill plate was strengthened by two layers of the GFRP laminateson each side and perpendicular to each other, as shown in Fig. 7-c.The hysteresis curves for the CSPSP4 are shown in Fig. 21.

In the elastic loading cycles, the panel behaved in a stiff manner.Elastic buckling occurred at low levels of loading at the drifts largerthan 0.75%. At the 7 mm deformation level, the shear strength of thespecimen was equal to 165.6 kN and after that it is not significantlychanged by the increasing deformations. Results show that in this

Fig. 11. Details of the SPSW1 (Alavi and Nateghi [42]).

-80 -60 -40 -20 0 20 40 60 80-1000

-500

0

500

1000

Inter Story Drift (mm)

Stor

y Sh

ear

(kN

)

ExperimentalNumerical (Cyclic)Numerical (Push-over)

Fig. 12. Comparative results between numerical and experimental models—goodagreement.



specimen due to the strengthening of the steel infill plate by two layersof the GFRP layer, which are perpendicular to each other, during thetesting the delamination and tearing phenomena were not observed(as shown in Fig. 22). Yet some debonding between the GFRP layersand the steel plate was observed.

3.5.5. CSPSP5 specimenIn the CSPSP5 specimen, principal orientation of the GFRP lami-

nates is oriented in the direction of the tension fields as given in

Fig. 7-d. The hysteresis curves for the specimen are shown inFig. 23. In the early loading cycles, the panel behaved in a stiff man-ner. Local buckling after deformations larger than 0.7% drift levelwas observed. Behavior of the specimen until 5 mm deformationlevel was linear. At the 5 mm deformation level, the shear strengthof the specimen was obtained to be 160 kN. At the larger deforma-tions, stiffness of the specimen was decreased. The specimen behavedstiff up to the 12 mm drift level (1.7% drift). At this deformation, theshear strength of the specimen was obtained to be 180.5 kN and tear-ing and delamination were not observed in this specimen. At largerdeformations, due to ruptures in the GFRP laminated layer and jointfailure as shown in Fig. 24, suddenly, shear strength of the specimenwas reduced.

a) SPSP1 b) CSPSP3

0

50

100

150

200

0 5 10 15

Forc

e (k

N)

Dispacement (mm)

SPSP1

0

50

100

150

200

0 5 10 15

Forc

e (k

N)

Dispacement (mm)

CSPSP3

Fig. 13. Push-over results of the SPSP1 and CSPSP3 specimens.

Fig. 14. Principal stress of the SPSP1 and CSPSP3 specimens (Pa.).

-20 -15 -10 -5 0 5 10 15 20-200

-150

-100

-50

0

50

100

150

200

Displacement (mm)

For

ce (

kN)

Fig. 15. Hysteresis curves of the SPSP1 specimen. Fig. 16. Local tears in the infill plate.



3.5.6. Major characteristics obtained during testing

3.5.6.1. Stiffness, shear strength. Stiffness and strength are the majorcharacteristics of the lateral resistant systems in resisting the earth-quake loads. In Table 4, stiffness, yield strength, ultimate strength,and ductility of the all specimens based on experimental hysteresiscurves are presented.

In this study, component stiffness or effective stiffness was calcu-lated based on the secant stiffness at yield level forces. Yield strengthis effective yield strength and was calculated based on the FEMA 356standard. These results show that by strengthening the steel infillplate with the GFRP laminated layers, the initial stiffness of the spec-imen was increased up to 59%. In the CSPSP2 and CSPSP3 specimensthat steel infill plates were strengthened by two layers of the GFRP,initial stiffness was increased by 10% and 39%, respectively. In theCSPSP4 and CSPSP5 specimens that steel infill plates are strengthenedby four layers of GFRP, initial stiffness was increased by 34% and 59%,respectively. These results show that the orientation of the GFRP lam-inates has a significant effect on the stiffness of the specimens. Ac-cordingly, if the principal orientation of the GFRP laminates isoriented in the direction of the tension fields, the stiffness in speci-mens would be the maximum possible value. In addition, changesin the direction of the GFRP laminates can change up to 25% in thestiffness of the specimen. Yield and ultimate strength of the all spec-imens are shown in Table 4. Results show both yield and ultimatestrength in the all strengthened specimens are significantly increased.Ultimate strength in the CSPSP2, CSPSP3, CSPSP4, and CSPSP5 speci-mens were increased by 24%, 71%, 101%, and 120%, respectively.

Therefore, concluding that the fiber orientation is an important vari-able in the ultimate strength of the C-SPSWs. The structural capacityof FRP laminates can be tailored and maximized by aligning the fibersalong the optimal orientation. For C-SPSWs, it is well established thatfibers should be aligned along the direction of the tension field. In thisstate, maximum stiffness and strength are provided by FRP laminates.

3.5.6.2. Ductility. Ductility can be defined as the ability of materials toundergo large deformations without rupture before failure. The duc-tility is calculated as:

μ ¼ Δy

Δmax

Where Δy and Δmax represent the displacement at the point of sig-nificant yielding and the maximum plastic displacement that the sys-tem sustains the loads up to the failure, respectively. Due tolimitations in testing set-up and experimental equipment ductilitywas calculated at almost 2% in most tests. Therefore, ductility was con-sidered only for this range. Ductility of the all specimen is shown inTable 4. Results indicate that the ductility in the specimens withstrengthened steel infill plate by one layer of GFRP in each side(CSPSP2 and CSPSP3) is decreased. Unlike the CSPSP2 and CSPSP3 spec-imens, the CSPSP4 specimen however, showed ductile behavior. Also,because of joint failure, the CSPSP5 specimen showed low-ductilitybehavior.

3.5.6.3. Secant stiffness. In Fig. 25, the comparison in terms of secantstiffness versus drift of the specimens is shown. The secant stiffnesswas calculated based on the tangent stiffness of the last cycle ofeach displacement increment as shown in Appendix A. In order tocontrol the story drift, secant stiffness is one of the important seismicparameters. Secant stiffness was decreased in the specimens alongwith increasing overall drifts. Secant stiffness in the all strengthenedspecimens was larger than the un-strengthened specimen (SPSP1).These results show that the orientation of the GFRP laminates has asignificant effect on the secant stiffness of the specimen. At the 2%drift, secant stiffness of the CSPSP2, CSPSP3, CSPSP4, and CSPSP5specimens increased by 48%, 81%, 100%, and 160%, respectively,concluding that the fiber orientation is an important variable in thesecant stiffness of the C-SPSW as well. Accordingly, if principal orien-tation of GFRP laminates is oriented in the direction of tension fields,the secant stiffness in specimens reaches the maximum possiblevalue.

-20 -15 -10 -5 0 5 10 15 20-200

-150

-100

-50

0

50

100

150

200

Displacement (mm)

For

ce (

kN)

Fig. 17. Hysteresis curves of the CSPSP2 specimen.

Delaminationand debonding

Fig. 18. Local tears and delamination in the GFRP layers (in the CSPSP2 specimen).

-20 -15 -10 -5 0 5 10 15 20-200

-150

-100

-50

0

50

100

150

200

Displacement (mm)

For

ce (

kN)




3.5.6.4. Cumulative dissipated energy. Cumulative hysteretic dissipatedenergy is the summation of the dissipated energy experienced by thespecimen during the test. In Fig. 26, the comparison in terms of cumu-lative (hysteretic) dissipated energy versus drift of all the specimensis shown. This parameter is one of the most important characteristicsaffecting the seismic performance of the C-SPSW system. In the spec-imens, with increasing drift, cumulative dissipated energy of thespecimens increased. Cumulative dissipated energy in the strength-ened specimens is larger than the un-strengthened specimen(SPSP1). The cumulative dissipated energy of the CSPSP2, CSPSP3,CSPSP4, and CSPSP5 specimens, at the almost 2% drift in which thewhole spectrum is considered, increased by 24%, 28%, 48%, and 50%,respectively. As it can be observed, the amount of the absorbed ener-gy in the CSPSP2 and CSPSP3 and also in the CSPSP4 and CSPSP5 isclose to each other. However, the cumulative dissipated energy inthe specimens was a little more, if the principal orientation of theGFRP laminates is oriented in the direction of the tension fields.Therefore, according to the results, fiber orientation is not a substan-tial variable in the cumulative dissipated energy of the C-SPSW.

3.5.6.5. Equivalent viscous damping ratio. The classical expression isconsidered for the assessment of the viscous damping coefficient ξeq:

ξeq ¼ ξ0 þ ξhyst

Where ξ0 corresponds to the initial damping in the elastic rangeand ξhyst corresponds to the equivalent viscous damping ratio thatrepresents the dissipation due to the nonlinear (hysteric) behavior.The method of calculating ξhyst is presented in Appendix A. InFig. 27, the comparison in terms of equivalent viscous damping ratioversus drifts of the specimens is provided. The results show thatwhile specimens were entered into the nonlinear region, the amountof equivalent viscous damping ratio in the SPSP1 and CSPSP2 speci-mens was approximately equal to 15%. The amount of this coefficientfor the CSPSP3 and CSPSP4 specimens was nearly equal to 12% and forthe CSPSP5 specimen was nearly equal to 8%. Accordingly, equivalentviscous damping ratio in the strengthened specimens, except theCSPSP2, is less than the un-strengthened specimen (SPSP1). Resultsalso show that if principal orientation of the GFRP layers lies in the di-rection of the tension fields, equivalent viscous damping ratio of thecomposite SPSP will decreased.

-20 -15 -10 -5 0 5 10 15 20-200

-150

-100

-50

0

50

100

150

200

Displacement (millimeter)

For

ce (

kN)


Fig. 20. Local tearing and delamination in the GFRP layers (in the CSPSP3 specimen).

Fig. 22. CSPSP4 specimen at the 2% drift.



3.5.6.6. Failure modes in the strengthened steel plates by GFRP laminatessubjected to shear loading. The results of this research indicate thatpossible failure modes in the strengthened steel plates by GFRPlaminates subjected to shear loading can be classified as:

(1) Steel and adhesive interface failure(2) GFRP layer delamination (separation of glass fibers from the

resin matrix)(3) Rupture in the main direction of GFRP layers(4) Rupture in the direction perpendicular to the main direction of

the GFRP laminate(5) Joints failure(6) Steel yielding; ( ductile mode)

It was found that the failure modes depend on the fiber orientationand the number of GFRP laminated layers. However, results show thatthe most common failure mode for FRP–strengthened steel platesubjected to the shear loading occurs at the steel and adhesive interface

and failure mode can be a combination of several failure modes. Thefailure modes in the specimens can be summarized as follows:

a Strengthened specimens by one layer of GFRP in each side(CSPSP2 and CSPSP3)

• Rupture in the direction perpendicular to the main direction ofthe GFRP laminate (common failure mode)

• GFRP layer delamination (separation of glass fibers from theresin matrix).

• Steel and adhesive interface failure.• Steel yielding

b Strengthened specimens by two layer of GFRP in each side(CSPSP4 and CSPSP5)

• Steel and adhesive interface failure.• Joints failure (only in the CSPSP5 specimen).

-20 -15 -10 -5 0 5 10 15 20-200

-150

-100

-50

0

50

100

150

200

Displacement (mm)

For

ce (

kN)


Fig. 24. Tearing and joint failure in the infill plate (in the CSPSP5 specimen).

Table 4Comparison of the stiffness and shear strength of Experimental models.

Specimen Stiffness (K) K. rel Yield strength (Fy) Fy. rel Ultimate strength (Fu) Fu.rel Ductility

kN/m (kN/m)/(kN/m) kN kN/kN kN kN/kN μ=Δy/Δmax

SPSP1 26,155.0 1.00 72.88 1 82.58 1 >4.9CSPSP2 28,666.7 1.10 100.5 1.38 102.3 1.24 4.48CSPSP3 36,325.3 1.39 115 1.58 140.8 1.71 3.36CSPSP4 34,981.7 1.34 160.7 2.20 165.6 2.01 >5CSPSP5 41,652.4 1.59 160 2.20 182 2.20 2.77

.64 .86 1.07 1.28 1.5 1.71 1.93 2.250

5

10

15

20

25

30

35

40

Displacement (m/m %)

Seca

nt s

tiff

ness

(kN

/mm

)

SPSP1CSPSP2CSPSP3CSPSP4CSPSP5

Fig. 25. Secant stiffness of the specimens.

0.64 0.86 1.07 1.28 1.5 1.71 1.93 2.250

0.5

1

1.5

2

2.5

3

3.5x 10

4


Cum

ulat

ive

Dis

sipa

ted

Ene

rgy

(kN

.m)

SPSP1CSPSP2CSPSP3CSPSP4CSPSP5

Fig. 26. Cumulative dissipated energy of the specimens.



• Rupture in the main direction of GFRP layers (only in the CSPSP5specimen).

• Steel yielding

4. Conclusion

In this paper, the nonlinear behavior of the composite steel plateshear wall systems, in which steel infill plate was strengthened by theGFRP laminated layers, was experimentally investigated. Tests weredesigned to evaluate the effect of the GFRP layers, the number of GFRPlayers and the orientation of GFRP layers on the stiffness, shear strength,cumulative dissipated energy and other important seismic parametersin the composite steel plate shear walls. The main results of this exper-imental study can be summarized as follows:

(1) If steel infill plate strengthened by GFRP layers, yield and ulti-mate strength of the C-SPSW will significantly increase. Fiberorientation is an important variable in the shear strength. Ifthe principal orientation of GFRP laminates is oriented in thedirection of the tension fields, the shear strength will increase.

(2) If steel infill plate strengthened by GFRP layers, the initial andsecant stiffness of the C-SPSW will significantly increase. Andalso, the initial and secant stiffness will increase, if the princi-pal orientation of the GFRP laminates is oriented in the direc-tion of the tension fields.

(3) Cumulative dissipated energy in the strengthened specimens islarger than the un-strengthened specimen. Fiber orientation isnot a substantial variable in the cumulative dissipated energyof the C-SPSWs. However if the principal orientation of GFRP

laminates is oriented in the direction of the tension fields, the cu-mulative dissipated energy in specimens can be enhanced little.

(4) Equivalent viscous damping ratio in the strengthened specimensis less than the un-strengthened specimen. Results show that ifprincipal orientations of the GFRP layers lie in the direction ofthe tension fields, the equivalent viscous damping ratio of thecomposite SPSPs will decrease.

(5) Test results show that the strengthened steel infill plate hasnegligible compression strength and as a result shear bucklingoccurs at the low levels of the loading like unstiffened specimen.Also, FRP laminated layers are effective in increasing post‐bucklingstrength due to the tension field action and also provide additionalsecant stiffness in the system.

(6) If steel infill plate strengthens by one layer of GFRP at each side,failure modes of the specimens will be rupture and delaminationof the GFRP laminates along the direction perpendicular to themain direction of the GFRP laminates.

(7) Failure modes depend on the fiber orientation and the number ofGFRP laminated layer. The most common failure mode occurs insteel and adhesive interface, and failure can be a combination ofseveral failure modes.

(8) Results indicate that the ductility in the specimens with strength-ened steel infill plate by GFRP layers can result in lower values.

(9) Fiber orientation is an important variable on the behavior of theC-SPSWs.

Acknowledgments

The authors wish to acknowledge the support of the IIEES; and SikaParsianCo. (supplier of the SikaDur 330 epoxy andGFRP laminate). Spe-cial thanks go to the IIEES structures laboratory personnel; Dr Ziyaeifarand technicians for their valuable contributions.

Appendix A. [41]

a- Secant stiffness (based on Fig. 28):

K sec ¼ΔFΔL

¼ Fmax−Fmin

Δmax−Δminð3Þ

b- Equivalent viscous damping ratio ζhyst (based on Fig. 28):

ζhyst ¼14π

EDES

� �ð4Þ

ED: Dissipated energy in each cycle that is equal to area of loop ineach cycle (see Fig. 28):

Es ¼ 18

Fmax−Fminð Þ � Δmax−Δminð Þ ð5Þ

References

[1] Astaneh-Asl A. Seismic behaviour and design of steel shear walls. Steel TIPS Re-port. Moraga, CA: Structural Steel Educational Council; 2001.

[2] Sabelli R, Bruneau M. Design guide 20: steel plate shear walls. Chicago, IL, USA:American Institute of Steel Construction; 2007.

[3] Haghani R. Analysis of adhesive joints used to bond FRP laminates to steelmembers—anumerical and experimental study. Construction and buildingmaterials 2010;24(11):2243–51.

[4] Zhao X-L, Zhang L. State-of-the-art review on FRP strengthened steel structures.Eng Struct 2007;29(8):1808–23.

[5] Khazaei-Poul M. Behavior of strengthened steel plate shear wall by FRP laminate.M.Sc. Thesis. International Institute Of Earthquake Engineering and Seismology,Iran. December; 2011.

[6] Wagner H. Flat sheet metal girders with very thin webs. Part i. General theoriesand assumptions. Tech Memo. National Advisory Committee for Aeronautics,Washington D. C., 1931, No. 604.Fig. 28. One loop of the hysteresis curves.

0.64 0.86 1.07 1.28 1.5 1.71 1.93 2.250

0.05

0.1

0.15

0.2


Equ

ival

ent v

isco

us d

ampi

ng fa

ctor SPSP1

CSPSP2CSPSP3CSPSP4CSPSP5

Fig. 27. Equivalent viscous damping ratio of the specimens.



[7] Thorburn LJ, Kulak GL, Montgomery CJ. Analysis of steel plate shear walls. Struc-tural Engineering Report No. 107, Department of Civil Engineering, University ofAlberta Edmonton, Alberta, Canada; 1983.

[8] Timler PA, Kulak GL. Experimental study of steel plate shear walls. StructuralEngineering Report, No. 114. Canada: University of Alberta; 1983.

[9] Elgaaly M. Thin steel plate shear walls behavior and analysis. Thin wall Struct1998;32:151–80.

[10] Berman JW, Bruneau M. Plastic analysis and design of steel plate shear walls.J Struct Eng 2003;129(11):1148–56.

[11] CAN/CSA S16. Limit state design of steel structures. Willowdale, ON., Canada:Canadian Standards Association; 2001.

[12] AISC. Specifications for structural steel buildings. Chicago, IL., USA: American Instituteof Steel Construction; 2005.

[13] Sabouri-Ghomi S, Ventura C, Kharrazi HK. Shear analysis and design of ductilesteel plate walls. J Struct Eng Jume 1 2005;131(6).

[14] Kharrazi MHK. Rational method for analysis and design of steel plate walls. Ph.D.dissertation. Vancouver (Canada): University of British Columbia; 2005.

[15] Khazaei-Poul M, Nateghi-Alahi F. Theoretical and numerical study on the strength-ened steel plate shear walls by FRP laminates. Int J Eng Trans C March 2012;25(1).

[16] Astaneh-Asl A. Seismic studies of innovative and traditional composite shearwalls. Research project in-progress. Berkeley: Dept of Civil and Env Engineering:Univ. of California; 1998–2000.

[17] Astaneh-Asl A. Cyclic tests of steel shear walls. Research project. Berkeley: Dept ofCivil and Env Engineering, Univ of California; 2001.

[18] Rahai A, Hatami F. Evaluation of composite shear wall behavior under cyclic load-ings. J Constr Steel Res 2009;65(7):1528–37.

[19] Lubell AS, Prion HGL, Ventura CE, Rezai M. Unstiffened steel plate shear wall per-formance under cyclic loading. J Struct Engin 2000;126(4):453–60.

[20] Caccese V, Elgaaly M, Chen R. Experimental study of thin steel-plate shear wallsunder cyclic load. ASCE J Struct Eng 1993;119(2):573–87.

[21] Driver RG, Kulak GL, Kennedy DJL. Cyclic tests of four-story steel plate shear wall.ASCE J Struct Eng 1998;124(2):112–20.

[22] Sabouri-Ghomi S, Roberts TM. Nonlinear dynamic analysis of steel plate shearwalls including shear and bending deformations. Eng Struct 1992;14(5):309–17.

[23] Vian D, Bruneau M, Tsai KC, Lin YC. Special perforated steel plate shear walls withreduced beam section anchor beams. I: experimental investigation. J Struct EngMarch 1 2009;135(3).

[24] Alinia MM. A study into optimization of stiffeners in plates subjected to in-planeshear loads. Thin Wall Struct 2005;43(4):845–60.

[25] Alinia MM, Dastfan M. Behaviour of thin steel plate shear walls regarding framemembers. J Constr Steel Res 2006;62:730–8.

[26] Cadei JMC, Stratford TJ, Hollaway LC, Duckett WG. Strengthening metallic struc-tures using externally bonded fibre reinforced composites. London: CIRIA; 2004.

[27] Wright PNH, Wu Y, Gibson AG. Fiber reinforced composite-steel connections fortransverse ship bulkheads. Plastic Rubber Compos 2000;29(10):549–57.

[28] Ekiz E, El-Tawil S. Inhibiting steel brace buckling using CFRP wraps. Proceedingsof the eighth national conference on EQ Engineering, San Francisco; 2006.

[29] Sen R, Liby L, Mullins G. Strengthening steel bridge sections using CFRP laminates.Compos Part B- Eng 2001;32(4):309–22.

[30] Moy SSJ, Barnes F, Moriarty J. Structural upgrading and life extension of struts andbeams using carbon fiber reinforced composite. Proceedings of the conference oncomposites and plastics in construction. Watford, UK: Building Research Estab-lishment; 1999. Paper 15:1–Paper 15:9.

[31] Deng Jun, Marcus MK, Lee B. Behaviour under static loading of metallic beamsreinforced with a bonded CFRP plate. Compos Struct 2007;78:232–42.

[32] Benachour A, Benyoucef S, Tounsi A, Adda-bedia EA. Interfacial stress analysis ofsteel beams reinforced with bonded prestressed FRP plate. Eng Struct November2008;30(11):3305–15.

[33] Colombi P, Bassetti A, Nussbaumer A. Analysis of cracked steel members reinforcedby pre-stress composite patch. Fatigue Fract Eng Mater Struct 2003;26:59–66.

[34] Jones SC, Civjan SA. Application of fibre reinforced polymer overlays to extendsteel fatigue life. J Compos Constr, ASCE 2003;7(4):331–8.

[35] Tavakkolizadeh M, Saadatmanesh H. Fatigue strength of steel girders strengthenedwith carbon fiber reinforced polymer patch. J Struct Eng, ASCE 2003;129(2):186–96.

[36] Miller TC, Chajes MJ, Mertz DR, Hastings JN. Strengthening of a steel bridge girderusing CFRP plates. ASCE J Bridge Eng 2001;6(6):514–22.

[37] Hollaway LC, Cadei J. Progress in the technique of upgrading metallic structureswith advanced polymer composites. Prog Struct Engng Mater 2002;4:131–48,doi:10.1002/pse.112.

[38] Photiou NK, Hollaway LC, Chryssanthopoulos MK. Strengthening of an artificiallydegraded steel beam utilising a carbon/glass composites system. Proceedings ofthe 2nd international conference on advances polymer composites for structuralapplications in construction. Guildford, UK: Woodhead Publishing Limited; 2002.

[39] ASTM, A-370–03 Standard test methods and definitions for mechanical testing ofsteel products. ASTM International, 100 Barr Harbor Drive, PO Box C700, WestConshohocken, PA 19428–2959, United States; 2003.

[40] ATC. Guidelines for cyclic seismic testing of components of steel structures for build-ings. Rep. No. ATC-24. Redwood City, Calif: Applied Technology Council; 1992.

[41] Craig Roy R, Jr Andrew J. Fundamentals of structural dynamics. Published by2nd edi-tion. Hoboken, New Jersey: JohnWiley & Sons, Inc; 2006. ISBN13: 978-0-471-43044-5.

[42] Alavi E, Nateghi F. Experimental and analytical study on boundary elementseffects on steel plate shear walls behavior. 12th European Conference on Earth-quake Engineering, paper No.45; 2010.


Experimental study of steel plate shear walls with infill plates strengthened by GFRP laminates

Documents

Transcript of Experimental study of steel plate shear walls with infill plates strengthened by GFRP laminates